Abstract There is growing interest in targeting DNA damage repair (DDR), however predictive markers are largely lacking and optimal combinations with targeted therapies have not been elucidated. In preliminary studies, we have shown that poly (ADP-ribose) polymerase (PARP) inhibitors have antitumor efficacy not only in PDX models with germline BRCA mutations, but also in PDX models with other germline or somatic alterations in DDR genes. Further, we have shown that K-Ras mutant cell lines are resistant to PARP inhibitors and that MEK inhibitors enhance the antitumor efficacy of PARP inhibitors. In other work, we have found that there is a significant enrichment of PI3K pathway alterations in patients with mutations in DDR genes (p=0.008). We propose that clinically and molecularly annotated PDX models can help identify predictive markers of response to DDR inhibitors and can be used to develop rational combination therapies. Our long term goal is to use molecular features of each patient?s tumor to optimize therapy selection. As a PDX development and trial center, we expect to build a large panel of PDXs to facilitate genotype (and other molecular subtype)- phenotype correlation. We hypothesize that tumors with DDR defects will be more likely to benefit from DNA damage inhibitors (such as PARP, ATR, Wee1), and that targeting actionable genomic co-alterations and adaptive responses may enhance anti-tumor efficacy.